Binaphthyl-based dicationic peptoids with therapeutic potential

Article abstract of DOI:10.1002/anie.200904392

(Figure Presented) Superbugs stalled! Two newly designed synthetic dicationic peptoids (see one example; red O, blue N, green Cl) show promising in vitro bactericidal activity against a range of Gram-positive pathogens, including organisms resistant to vancomycin, methicillin, and linezolid, with only slow development of resistance. Moreover their potency is maintained in vivo.

Full text of DOI:10.1002/anie.200904392

Angewandte
Chemie
DOI: 10.1002/anie.200904392
Antibiotics
Binaphthyl-Based Dicationic Peptoids with Therapeutic Potential**
John B. Bremner,* Paul A. Keller,* Stephen G. Pyne,* Timothy P. Boyle, Zinka Brkic,
Dorothy M. David, Adel Garas, Jody Morgan, Mark Robertson, Kittiya Somphol,
Michael H. Miller, Adam S. Howe, Paul Ambrose, Sujata Bhavnani, Thomas R. Fritsche,
Douglas J. Biedenbach, Ronald N. Jones, Robert W. Buckheit, Jr., Karen M. Watson,
Dean Baylis, Jonathan A. Coates, John Deadman, Dharshini Jeevarajah, Andrea McCracken,
and David I. Rhodes
While the cationic glycopeptide vancomycin has long been
regarded as the gold standard for the treatment of recalcitrant
Gram-positive bacterial infection, this position has been
compromised by the emergence of resistant strains.[1–3] The
first report[4] of such resistance emerged in 1988, and has
subsequently widened amongst the enterococci[5] and staph-
ylococci, including methicillin-resistant Staphylococcus
aureus (MRSA);[1,2] cross-resistance to linezolid[6] is also a
concern. Some recent chemical strategies for overcoming this
resistance[7] have centered on other high molecular weight
cyclic peptides,[8–10] elegantly crafted vancomycin[11] or vanco-
mycin aglycone[12,13] analogues, potent dual-action vancomy-
cin/b-lactam hybrid antibiotics,[14] or large vancomycin
dimers.[15,16] An alternative strategy is to design smaller,
simpler cationic peptoids with some related design features to
vancomycin which could still interact with the altered
peptide-glycan cell-wall moiety in both vancomycin-resist-
ant[3] and -sensitive strains and thus broaden the antibacterial
spectrum. Svendsen et al. designed minimal cationic peptido-
mimetics, and a pharmacophore has been developed for
dipeptides which includes the presence of two cationic
charges and two hydrophobic units of steric bulk.[17–19]
Subsequently, cationic tripeptide analogues were devel-
oped[20,21] that demonstrated good activity against both
Gram-positive (including MRSA) and Gram-negative bac-
teria, but were not evaluated with respect to vancomycin-
resistant strains.[20,21]
In our minimization strategy we explored cyclic peptoids
containing a hydrophobic scaffold[22,23] that specifically con-
tained 1) a basic amino acid residue to potentially engage in
an ionic interaction with the terminal carboxylate group of
(R)-Ala or (R)-Lac in the bacterial cell-wall peptide-glycan
unit, 2) a tripeptide moiety for H-bonding interactions with
either of these two terminal groups, and 3) a 1,1’-binaphthyl
system (or related systems[24]) to enable hydrophobic inter-
actions with the (R)-Ala methyl group or with other hydro-
phobic regions. Subsequently, through selective manipula-
tions of the C terminus and addition of a further basic amino
acid component, the Lys-Arg-containing acyclic peptoid 1
(Scheme 1; unpublished work) was found with a minimum
inhibitory concentration (MIC) of 2 mgmLÀ1 against S. aureus,
but 31 mgmLÀ1 against vancomycin-resistant enterococci
(VRE). Modification of the 2-naphthyl ether group and the
C-terminal region then resulted in ester 2a and the ester-
bioisosteric oxazole 2b (Scheme 1), which are the subject of
this paper. These two dicationic peptoids represent a promis-
ing new antibacterial subclass of the cationic peptides.
A modular approach amenable to scaleup was adopted for
the synthesis of 2a and 2b (Scheme 2). The key precursor for
the scaffold was commercially available (S)-1,1’-binaphtha-
lene-2,2’-diol (3). Base-induced mono-O-alkylation of 3 with
1-bromo-3-methylbutane, followed by O-alkylation of the
remaining phenolic group with bromoacetic acid together
with K2CO3, then provided access to key acid 4. Compound 4
was then coupled with H-(R)-Lys(Boc)-OMe to give pro-
tected peptoid 5, and subsequently acid 6 by base hydrolysis
followed by acidification. N-Protected precursors of peptide
amine salts 2a and 2b were accessed by diimide-mediated
coupling of the appropriate starting amines 7 with an amine/
amidine-group-protected (R)-arginine to give 8a,b, followed
by Fmoc deprotection to afford the amine intermediates 9a,b
in good yield. The EDCI/HOBT-promoted coupling of
amines 9a,b with acid 6 then gave the respective penultimate
amide-coupled products (84% yield for the isopentyl ester
[*] Prof. J. B. Bremner, Assoc. Prof. P. A. Keller, Prof. S. G. Pyne,
Dr. T. P. Boyle, Dr. Z. Brkic, D. M. David, Dr. A. Garas, Dr. J. Morgan,
Dr. M. Robertson, Dr. K. Somphol
School of Chemistry, University of Wollongong
Wollongong, NSW 2522 (Australia)
Fax: (+61)2-42214287
E-mail: john_bremner@uow.edu.au
keller@uow.edu.au
spyne@uow.edu.au
D. Baylis, Dr. J. A. Coates, Dr. J. Deadman, D. Jeevarajah,
Dr. A. McCracken, Dr. D. I. Rhodes
Avexa Ltd, 576 Swan Street, Richmond, Victoria 3121 (Australia)
Dr. M. H. Miller, Dr. A. S. Howe, Dr. P. Ambrose, Dr. S. Bhavnani
Institute for Clinical Pharmacodynamics, Ordway Research Institute
43 British American Blvd, Latham, NY (USA)
Dr. T. R. Fritsche, Dr. D. J. Biedenbach, Dr. R. N. Jones
JMI Laboratories, North Liberty, Iowa (USA).
Dr. R. W. Buckheit, Jr., K. M. Watson
ImQuest BioSciences, Inc. Frederick, MD 21704 (USA).
[**] We thank the National Health and Medical Research Council
(Development Grants 303415 and 404528), the Uni. of Wollongong,
and Avexa Ltd for financial support; R. Nation (Monash University)
for assistance with some testing against Gram-negative pathogens;
and Washington Biosciences, Inc, Maryland (USA) for the nasal
decolonisation and systemic studies.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200904392.
Angew. Chem. Int. Ed. 2010, 49, 537 –540
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
537

Communications
and 64% yield for the oxazole analogue). Deprotection of
these amide products with TFA, addition of hydrogen
chloride in diethyl ether, and freeze-drying gave the required
salts 2a and 2b (Scheme 2). This general synthetic method-
ology was also used to access all diastereomers of peptoid 2a
for structure–activity relationship studies (see the Supporting
Information). The structural and stereochemical integrity and
purity (> 95%) of 2a, its diastereomers, and 2b were
established by HPLC and NMR analysis.
Compounds 2a and 2b are active against a variety of
organisms, including S. aureus, coagulase-negative staphylo-
cocci, enterococci, and streptococci (Table 1 and the Support-
Table 1: In vitro antibacterial activity of 2a, 2b, and vancomycin (Van)
against Gram-positive and Gram-negative isolates.
Strain
MIC50 (range) or MIC [mgmLÀ1
]
(number of strains)
2a
2b
Van
S. aureus
MSSA (8)
4 (2–4)
4
4
4
4
4
4
4
1 (1–2)
2 (1–2)
8
>32
1
1
2
MRSA (7)[a]
4
4
2
2
4
4
Scheme 1. Structures of dicationic peptoids 1 and 2a/b.
VISA (1)[b]
VRSA (1)[b]
PVL [CA-MRSA] (1)[b]
Linezolid R (1)[b]
S. epidermidis (3)[a]
E. faecium
VRE (4)[a]
4
4
>32
VSE (4)
Streptococci (13)
2 (2–4)
8 (4–16)
4
4 (2–8)
1 (0.5–1)
8 (4–16)
Gram-negative bacilli
E. coli (1)[b]
16
>32
4 (4 to >32)
8 (4–8)
32
>32
8 (2 to >32)
16 (4–16)
>32
n.t.
n.t.
K. pneumoniae (3)[a]
A. baumannii (4)
S. maltophilia (3)
n.t.
[a] Where the MIC was the same for each strain, no range is given.
[b] Where only one strain was tested, the value given is an MIC. S.
epidermidis=Staphylococcus epidermidis; E. faecium=Enterococcus fae-
cium; E. coli=Escherichia coli; K. pneumoniae=Klebsiella pneumoniae; A.
baumannii=Acinetobacter baumannii; S. maltophilia=Stenotrophomonas
maltophilia. MSSA=methicillin-sensitive S. aureus; MRSA=methicillin-
resistant S.aureus; VISA=vancomycin-intermediate S. aureus; VRSA=
vancomycin-resistant S. aureus; PVL=Panton–Valentine Leucocidin-
positive S. aureus (CA-MRSA=community-acquired MRSA); Linezo-
lid R=linezolid-resistant; VRE=vancomycin-resistant enterococci;
VSE=vancomycin-sensitive enterococci; n.t.=not tested.
Scheme 2. General synthetic route to 2a and 2b. (Note, in the
compound numbers below, “a” corresponds to the R group in 2a and
“b” to the R group in 2b). a) 1-Bromo-3-methylbutane, K2CO3,
CH3COCH3, reflux, 16 h; b) BrCH2COOH, K2CO3, MeOH, reflux, 3 h,
83% (2 steps); c) H-(R)-Lys(Boc)-OMe·HCl, EDCI, HOBt, MeCN, RT,
3 h, 90%; d) 4% LiOH/H2O, THF, RT, 16 h, 89%; e) Fmoc-(R)-
Arg(PG)-OH, EDCI, HOBt, MeCN, RT, 3 h, 8a (97%), 8b (94%);
f) piperidine, MeCN, RT, 3 h, 9a (91%), 9b (97%); g) EDCI, HOBt,
MeCN, RT, 3 h; h) TFA, CH2Cl2, RT, 16 h; then 1m HCl/Et2O, 2a
(92%), 2b (62%). Boc=tert-butoxycarbonyl, EDCI=1-(3-dimethylami-
nopropyl)-3-ethylcarbodiimide hydrochloride, Fmoc=9-fluorenylme-
thoxycarbonyl, HOBt=N-hydroxybenzotriazole, Pmc=2,2,5,7,8-pen-
tamethyl-3,4-dihydro-2H-1-benzopyran-6-sulfonyl, TFA=trifluoroacetic
acid.
ing Information). The test compounds had a broad spectrum
of activity against the Gram-positive bacteria, with MIC50 (or
MIC) generally in the range 2–8 mgmLÀ1, but reduced activity
against streptococci with an MIC50 range of 4–16 mgmLÀ1 and
poor activity against Gram-negative bacteria (8 to
> 32 mgmLÀ1). The test compounds were also active against
organisms resistant to vancomycin, methicillin, and linezolid,
as well as community-acquired MRSA.
In further biological testing, 2a and 2b showed rapid
killing of MRSA within the first hour at concentrations of
ꢀ 2 ꢀ MIC (Figure 1a and b).[25] The rate of killing was dose-
dependent and faster and greater in magnitude than that
538
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 537 –540

Angewandte
Chemie
(i.p.) twice a day for 4 d; initial oral infection[26] with 5 ꢀ 1010
colony-forming units (cfu) per mouse, n = 10) effectively
reduced the number of viable bacteria remaining in the
spleens (11 cfu per spleen); this compared favorably with the
result for vancomycin (6 cfu per spleen), while with the
control (DMSO vehicle), 128 cfu per spleen remained.
Similar results were obtained for 2b, vancomycin, and the
control for the other bacterial inocula tested (5 ꢀ 109 and 1.5 ꢀ
1010 cfu per mouse).
Activity was established topically by using the mouse
nasal decolonization model[27–28] with MRSA. Twice-daily
administration for 5 d of 2a or 2b resulted in a significant
dose-dependent reduction in MRSA nasal carriage. A single
administration of neat 2a (5 wt%, approximately equimolar
to 2% mupirocin) was as active as 2% mupirocin, the
currently used clinical antibiotic,[29] administered twice daily
for 5 d. Ester 2a may act as a prodrug that is hydrolyzed to the
corresponding carboxylic acid; separate synthesis (by LiOH-
mediated hydrolysis of 2a and acidification with HCl) and
assessment of this acid showed it was considerably less active
than 2a (S. aureus (ATCC6538P) MIC 25 mgmLÀ1). Ester 2a
was stable for up to 4 h in human and mouse plasma, and 19 h
in minipig plasma (unpublished results). In vitro resistance
selectivity studies showed that resistance to 2a develops
slowly in S. aureus. Some 50 generations were required for a
tenfold increase in MIC to be observed. In comparison,
vancomycin took 18 generations. These strains resistant to 2a
remained susceptible to vancomycin, and this suggests that
2a, at least in part, has a different mode (or modes) of action.
The peptoid salt 2a, with its ease of synthesis compared to
the oxazole 2b, coupled with reasonable plasma stability and
amenability to stable nonaqueous formulation, is suitable for
topical application and has no irritancy potential in rabbit skin
irritation studies (data not shown). It is being developed for
indications including nasal decolonization and/or wound
infection/catheter-related infections.
Our results suggest that more than one mode of action is
exhibited by compounds 2a and 2b. From ESI-MS studies, it
was clear that 2a could form complexes with the cell-wall
model peptide sequences Ala-Lys-Ala-(R)-Ala-(R)-Ala
(mimicking the sequence in vancomycin-sensitive S. aureus)
and Ala-Lys-Ala-(R)-Ala-(R)-Lac (as an analogue of the
sequence in vancomycin-resistant S. aureus) with signals for
the doubly charged 1:1 complex evident at m/z 642 and 643,
respectively, together with separate peaks for the individual
components in each case. In contrast, vancomycin only
complexed with the terminal (R)-Ala sequence (m/z 927).
This indicates that inhibition of cell-wall cross-linking through
peptide complexation could contribute to the mode of action.
All diastereomers of 2a were equipotent against S. aureus
(vancomycin-sensitive and -resistant strains) and, together
with activity of 2a against MRSA and linezolid-resistant S.
aureus, this suggests a more general activity component for
disruption of Gram-positive bacterial cell membranes. The
slow development of resistance and the rapid bactericidal
action of 2a and 2b is also consistent with such an action.
Controversy surrounds the precise events responsible for
bacterial death in vitro on exposure to other cationic peptides.
Evidence for either cytoplasmic membrane damage inducing
Figure 1. Mean time–kill values against MRSA for a) 2a, b) 2b, and
c) vancomycin.
observed with vancomycin (Figure 1a–c). Peptoids 2a and 2b
were rapidly bactericidal (at 2 ꢀ MIC) with an associated, but
not simultaneous, reduction in optical density suggesting the
occurrence of bacterial cell lysis. The degree of bacteriolysis
was concentration-dependent (not shown). The results indi-
cate that bacteriolysis follows loss of cell viability, and
possibly suggests a complex mode of bacterial killing by
these compounds.
The observed in vitro activity was translated into in vivo
activity both systemically and topically. Since oxazole 2b is
more resistant to enzymatic hydrolysis than the terminal ester
analogue, it was tested systemically in an animal model which
measured the bacterial growth of MRSA (vancomycin-
sensitive strain BAA-41) in the spleens of infected mice.
Compound 2b dosed systemically (5 mgkgÀ1 intraperitoneal
Angew. Chem. Int. Ed. 2010, 49, 537 –540
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
539

Communications
changes in permeability[30] or a combination of both mem-
[11] K. C. Nicolaou, S. Y. Cho, R. Hughes, N. Winnsinger, C.
Smethurst, H. Labischinski, R. Endermann, Chem. Eur. J.
2001, 7, 3798 – 3823.
[12] C. M. Crane, D. L. Boger, J. Med. Chem. 2009, 52, 1471 – 1476.
[13] B. M. Crowley, D. L. Boger, J. Am. Chem. Soc. 2006, 128, 2885 –
2892.
[14] D. D. Long, J. B. Aggen, B. G. Christensen, J. Kevin Judice, S. S.
Hegde, K. Kaniga, K. M. Krause, M. S. Linsell, E. J. Moran, J. L.
Pace, J. Antibiot. 2008, 61, 595 – 602.
[15] K. C. Nicolaou, R. Hughes, S. Y. Cho, N. Winnsinger, H.
Labischinski, R. Endermann, Chem. Eur. J. 2001, 7, 3824 – 3843.
[16] J. Lu, O. Yoshida, S. Hayashi, H. Arimoto, Chem. Commun.
2007, 251 – 253.
[17] M. B. Strøm, B. E. Haug, M. L. Skar, W. Stensen, T. Stiberg, J. S.
Svendsen, J. Med. Chem. 2003, 46, 1567 – 1570.
[18] B. E. Haug, W. Stensen, T. Stiberg, J. S. Svendsen, J. Med. Chem.
2004, 47, 4159 – 4162.
[19] B. E. Haug, W. Stensen, J. S. Svendsen, Bioorg. Med. Chem. Lett.
2007, 17, 2361 – 2364.
brane damage and interaction with other targets within the
cell has been reported.[31,32] This type of dual action has been
established for the vancomycin analogue telavancin, which
compromises cell-wall synthesis and cell-membrane coher-
ence.[10]
In conclusion, two novel dicationic peptoids have been
designed and synthesized. These compounds show promising
in vitro antibacterial potency against a range of Gram-
positive pathogens including strains of S. aureus resistant to
vancomycin, methicillin, and linezolid. Peptoids 2a and 2b
are rapidly bactericidal with a concentration-dependent
profile, and induced resistance has been shown to develop
exceptionally slowly in vitro. These compounds may act by
more than one mode of action including inhibition of cell-wall
cross-linking and cell-membrane disruption. Systemic and
topical in vivo potency was maintained in mouse models of
infection, and 2a is being developed for topical indications.
[20] B. E. Haug, W. Stensen, M. Kalaaji, O. Rekdal, J. S. Svendsen, J.
Med. Chem. 2008, 51, 4306 – 4314.
[21] K. Flemming, C. Klingenberg, J. P. Cavanagh, M. Sletteng, W.
Stensen, J. S. Svendsen, T. Flaegstad, J. Antimicrob. Chemother.
2009, 63, 136 – 145.
[22] J. B. Bremner, J. A. Coates, D. R. Coghlan, D. M. David, P. A.
Keller, S. G. Pyne, New J. Chem. 2002, 26, 1549 – 1551.
[23] A. Garas, J. B. Bremner, J. Coates, J. Deadman, P. A. Keller,
S. G. Pyne, D. I. Rhodes, Bioorg. Med. Chem. Lett. 2009, 19,
3010 – 3013.
[24] T. P. Boyle, J. B. Bremner, J. A. Coates, J. Deadman, P. A. Keller,
S. G. Pyne, D. I. Rhodes, Tetrahedron 2008, 64, 11270 – 11290.
[25] S. A. Shelburne, D. M. Musher, K. Hulten, H. Ceasar, M. Y. Lu,
I. Bhaila, R. J. Hamill, Antimicrob. Agents Chemother. 2004, 48,
4016 – 4019.
[26] For dissemination of S. aureus after oral administration in mice,
see D. J. Hess, R. M. Garni, M. J. Henry-Stanley, C. L. Wells,
Shock 2005, 24, 376 – 381.
[27] K. B. Kiser, J. M. Cantey-Kiser, J. C. Lee, Infect. Immun. 1999,
67, 5001 – 5006.
[28] M. S. Rouse, M. Rotger, K. E. Pipier, J. M. Steckelberg, M.
Scholz, J. Andrews, R. Patel, Antimicrob. Agents Chemother.
2005, 49, 3187 – 3191.
[29] The Merck Index, 14th ed. (Ed.: M. J. OꢂNeil), Merck & Co.,
Inc., Whitehouse Station, NJ, 2006, pp. 1089 – 1090.
[30] R. E. W. Hancock, D. S. Chapple, Antimicrob. Agents Chemo-
ther. 1999, 43, 1317 – 1323.
[31] M. Wu, E. Maier, R. Benz, R. E. W. Hancock, Biochemistry
1999, 38, 7235 – 7242.
Received: August 6, 2009
Revised: September 7, 2009
Published online: December 11, 2009
Keywords: antibiotics · biological activity · bioorganic chemistry ·
.
drug discovery · peptidomimetics
[1] D. P. Levine, South. Med. J. 2008, 101, 284 – 291.
[2] D. M. Sievert, J. T. Rudrik, J. B. Patel, L. C. McDonald, M. J.
Wilkins, J. C. Hageman, Clin. Infect. Dis. 2008, 46, 668 – 674.
[3] D. Kahne, C. Leimkuhler, W. Lu, C. Walsh, Chem. Rev. 2005,
105, 425 – 448.
[4] A. H. Uttley, C. H. Collins, J. Naidoo, R. C. George, Lancet 1988,
331, 57 – 58.
[5] A. H. Mutnick, D. J. Biedenbach, R. N. Jones, Diagn. Microbiol.
Infect. Dis. 2003, 46, 63 – 68.
[6] R. D. Gonzales, P. C. Schreckenberger, M. B. Graham, S. Kelkar,
K. Den Besten, J. P. Quinn, Lancet 2001, 357, 1179 – 1183.
[7] F. von Nussbaum, M. Brands, B. Hinzen, S. Weigand, D. Hꢁbich,
Angew. Chem. 2006, 118, 5194 – 5254; Angew. Chem. Int. Ed.
2006, 45, 5072 – 5129.
[8] S. J. Projan, P. A. Bradford, Curr. Opin. Microbiol. 2007, 10,
441 – 446.
[9] F. Van Bambeke, Curr. Opin. Pharmacol. 2004, 4, 471 – 478.
[10] D. L. Higgins, R. Chang, D. V. Debabov, J. Leung, T. Wu, K. M.
Krause, E. Sandvik, J. M. Hubbard, K. Kaniga, D. E.
Schmidt, Jr., Q. Gao, R. T. Cass, D. E. Karr, B. M. Benton, P. P.
Humphrey, Antimicrob. Agents Chemother. 2005, 49, 1127 –
1134.
[32] H. Jenssen, P. Hamill, R. E. W. Hancock, Clin. Microbiol. Rev.
2006, 19, 491 – 511.
540
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 537 –540